This item is available under a Creative Commons License for non-commercial use only
Computer Sciences, Information Science
Abstract An increasing number of dynamic software evolution approaches is com- monly based on integrating or utilising new pieces of software. This requires reso- lution of issues such as ensuring awareness of newly available software pieces and selection of most appropriate software pieces to use. Other chapters in this book dis- cuss dynamic software evolution focusing primarily on awareness, integration and utilisation of new software pieces, paying less attention on how selection among different software pieces is made. The selection issue is quite important since in the increasingly dynamic software world quite a few new software pieces occur over time, some of which being of lower utility, lower quality or even potentially harmful and malicious (for example, a new piece of software may contain hidden spyware or it may be a virus). In this chapter, we describe how computational trust and reputation can be used to avoid choosing new pieces of software that may be malicious or of lower quality. We start by describing computational models of trust and reputation and subsequently we apply them in two application domains. Firstly, in quality assessment of open source software, discussing the case where different trustors have different understandings of trust and trust estimation methods. Sec- ondly, in protection of open collaborative software, such as Wikipedia.
Seigneur,J, M., Dondio, P., Trust and Reputation for Successful Software Self-Organisation, in Di Marzo Serugendo, Giovanna, Marie-Pierre Gleizes, and Anthony Karageorgos.(eds) "Self-organising Software: From Natural to Artificial Adaptation." Springer (2011).